US8121235B1ActiveUtility
Dimension reduction for codebook search
Est. expiryApr 1, 2028(~1.7 yrs left)· nominal 20-yr term from priority
H04B 7/0456H04L 5/0023H04L 2025/03426H04L 25/03343H04L 1/20H04L 2025/03808
97
PatentIndex Score
50
Cited by
5
References
16
Claims
Abstract
Systems are provided for searching for a codeword from a plurality of codewords in a codebook for use in precoding, for example, as used in a multiple-input multiple-output (MIMO) transmission system. Dimension reduction techniques may be utilized to reduce the complexity and enhance the efficiency of the codebooks search. Null-spaces of an optimal codeword and codewords in a codebook may be computed. Distance values may be computed based on the null=spaces of the codewords. A codeword may be selected from the codebook based on a minimum distance value from the optimal codeword.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1. A system for searching for a codeword from a plurality of codewords in a codebook for use in precoding, the system comprising:
a memory capable of storing a codebook with a plurality of codewords and null-spaces of the plurality of codewords; and
a processor capable of:
determining an optimal codeword based on channel conditions;
determining a null-space of the optimal codeword;
computing distance values based on distances between the null-space of the optimal codeword and the null-spaces of the plurality of codewords; and
selecting a codeword from the codebook based on a minimum distance value from the optimal codeword.
2. The system of claim 1 , wherein determining the optimal codeword is further based on a characteristic to optimize.
3. The system of claim 2 , wherein the characteristic to optimize is selected from the group consisting of:
signal to noise ratio,
mean squared error,
mutual information,
error rate, or
outage.
4. The system of claim 1 , wherein the distance values are computed based on a metric selected from the group consisting of:
chordal distance,
Fubini-Study distance, or
projection 2-norm distance.
5. The system of claim 1 , wherein the null-space of the optimal codeword comprises columns N s +1 to N t of V full and N s is an integer, N t is an integer, and V full is a full-rank matrix associated with the optimal codeword.
6. The system of claim 1 , wherein the processor is further capable of reducing at least one dimension of the null-space of the optimal codeword and the null-spaces of the codewords in the codebook prior to computing the distance values.
7. The system of claim 1 , wherein the selected codeword is found according with the equation:
Codeword
opt
=
argmin
1
≤
k
≤
2
L
D
chor
⌊
Null
(
V
opt
)
,
Null
(
W
k
)
⌋
,
wherein D chor is the chordal distance, Null(V opt ) is the null-space of the optimal codeword, V opt , and Null(W k ) is the null-space of a kth codeword in the codebook.
8. A system for searching for a codeword from a plurality of codewords in a codebook for use in precoding, the system comprising:
a memory capable of storing a codebook with a plurality of codewords and dimension reduced versions of the plurality of codewords, wherein the dimension of the plurality of codewords is reduced by at least one dimension; and
a processor capable of:
determining an optimal codeword based on channel conditions;
reducing at least one dimension of the optimal codeword;
computing distance values based on distances between the dimension reduced optimal codeword and the dimension reduced plurality of codewords; and
selecting a codeword from the codebook based on a minimum distance value from the optimal codeword.
9. The system of claim 8 , wherein dimensions of the optimal codeword and dimensions of the plurality of codewords in the codebook are reduced by arbitrarily removing at least one column of the optimal codeword and the plurality of codewords.
10. The system of claim 8 , wherein determining the optimal codeword is further based on a characteristic to optimize.
11. The system of claim 10 , wherein the characteristic to optimize is selected from the group consisting of:
signal to noise ratio,
mean squared error,
mutual information,
error rate, or
outage.
12. The system of claim 8 , wherein the distance values are computed based on a metric selected from the group consisting of:
chordal distance,
Fubini-Study distance, or
projection 2-norm distance.
13. The system of claim 8 , wherein the processor is further capable of determining corresponding null-spaces of the dimension reduced optimal codeword and the dimension reduced plurality of codewords in the codebook prior to computing the distance values.
14. The system of claim 13 , wherein the null-space of the dimension reduced optimal codeword is comprised of columns removed to reduce the dimensions of the optimal codeword.
15. The system of claim 13 , wherein the null-space of the dimension reduced plurality of codewords in the codebook is comprised of columns removed to reduce the dimensions of the codewords in the codebook.
16. The system of claim 8 , wherein the selected codeword is found according with the equation:
Codeword
opt
=
argmax
1
≤
k
≤
2
L
{
max
1
≤
j
≤
N
t
V
opt
(
:
,
N
t
)
H
W
k
(
:
,
j
)
}
,
wherein L is an integer, N t is an integer, V opt is the optimal codeword, and W k is a kth codeword in the codebook.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.